You are not logged in.
I have a mech keyboard which uses an Arduino Pro Micro as the controller (Nyquist, for those curious). I was doing updates to it a couple of days ago no problem but immediately after updating the kernel it started erroring out when flashing saying that address was out of range. This is normally an error with the code so I investigated and was able to rule it out because when I try the exact same code on the same keyboard on my work mac, it updates no problem. So I went to a forum for this keyboard firmware (qmk) and another arch user there experienced the same problem at the same time as I did. They rolled back to 10-1 and that fixed it for them.
I didn't have a cached package so I first tried just installing a different kernel (linux-pf) but that didn't work. So then I cloned the arch repo, checked out to the commit when 10-1 was added to core, rebuilt and installed. Still no dice. Then I realized that there's a package archive page so, just in case, I downloaded 10-1 of linux and linux-headers and installed them. Again, still not working.
So, at this point my working theory is that something related to my usb modules is not quite right. USB works fine for standard communication. In fact I am using a different mech keyboard right now to type this (curious: Pok3r).
Thing is, I do not know enough about the kernel to know how to debug this and track down the root cause. Anybody else experiencing this issue? Can anybody point me in the right direction to figure this out? Thanks!
Offline
Welcome to the arch linux forums HokieGeek. 4.17.10-1 worked correctly, 4.17.11-1 has the issue. 4.17.10-1 downgrade the issue is still present?
You verified the booted kernel with uname is now 4.17.10-1? Were any other packages upgrade apart from the kernel?
Have you compared the kernel message from a 4.17.10 boot and a 4.17.11 boot?
Offline
Linux archie 4.17.10-1-ARCH #1 SMP PREEMPT Wed Jul 25 11:23:00 UTC 2018 x86_64 GNU/Linux
There were indeed a few other packages as it had been a good week since my last update. The only ones I really remember are haskell packages cause I had to reinstall them when I couldn't rebuild xmonad.
I have not done a comparison, no. That being said, which kernel message exactly? The only thing I've looked it is the messaging when I try to upload the hex:
[11162.389175] usb 1-1.2: USB disconnect, device number 9
[11162.629276] usb 1-1.2: new full-speed USB device number 10 using ehci-pci
[11162.727570] usb 1-1.2: New USB device found, idVendor=2341, idProduct=0037, bcdDevice= 0.01
[11162.727577] usb 1-1.2: New USB device strings: Mfr=2, Product=1, SerialNumber=0
[11162.727581] usb 1-1.2: Product: Arduino Micro
[11162.727585] usb 1-1.2: Manufacturer: Arduino LLC
[11162.728298] cdc_acm 1-1.2:1.0: ttyACM0: USB ACM device
EDIT: formatting
Last edited by HokieGeek (2018-08-02 13:30:07)
Offline
There are a couple of result, while searching 2341:0037 arduino. But I suspect there could be problem if your are using a Arduthing clone. There are several one which mounting CH340 and some times it won't even work since the first day.
Have you try with a different kernel, is that the primer cause ?
do it good first, it will be faster than do it twice the saint
Offline
The keyboards are using clones, so I tried to flash a genuine mini and it gave me the same error. I have tried linux-pf:4.17.11-1, linux:4.17.11-1, and linux:4.17.10-1. They all give me the same error. That isn't a large sample set though. Right now my main reason for thinking it is the kernel is based on:
1. It is the most notable change
2. It used to work and still works on my mac
3. Another user resolved the same problem on his arch box by rolling back to 10-1
Not particularly strong reasons, but it's all I have.
Offline
So you might need to set up a virtual machine with a working kernel
do it good first, it will be faster than do it twice the saint
Offline
That is certainly something to try
Offline
What kernel would be considered working though, would you have to have that kernel on the host or just the guest and PCI pass through to pass the usb controller to the guest?
Offline
Right, I was scratching my head about that too.
Offline
For anybody finding this, the issue was with avr-gcc 8.2. Downgrading to 8.1 got things working again.
Offline